Find the waste hiding in your team’s expenses.
An expense is not waste just because it is large, recurring or unfamiliar. Follow the money back to the work it supported, and see which charges survive that examination.
Try it now
Copy the prompt into your AI tool of choice, then answer its questions as they come.
Paste it into any AI chat to start. For most people, this is all you need.
Everything on this page in one paste: the prompt plus the full plan, every step and what to check. Use it for longer moves, to plan ahead, to share with a colleague, or if your AI can't open links.
How it goes
What the AI will do
- Inspect expenses, invoices, subscriptions and usage, then search messages and project records for the purpose and commitments behind the spending.
- Investigate bounded categories, verify promising findings, and use a separate reviewer where available to check the amounts and consequences.
What to have ready
- The team, time period and any known spending question.
- Access to expense records, invoices, subscriptions, usage and relevant project communications; files or usable exports also work.
Where you come in
- Clarify material gaps about use, commitments or service needs that the records cannot resolve.
- Review the itemized findings and choose whether any proposed change should proceed separately.
What you'll get
An itemized expense review showing verified problems, unresolved candidates, supported savings estimates and what each proposed change would affect; no purchases or cancellations are made.
You're done when
You have a checked itemized review, with verified amounts, unresolved leads, realistic savings calculations and the effects of proposed changes clearly shown. The review method and any limits are stated.
These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result. ### Start with the activity and the user's context Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear. Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them. ### Check the capabilities needed for this work Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit. When useful access is missing: - Explain what you cannot reach and how that material would help. - Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot. - If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response. - Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available. ### Ask focused questions and respect the answers Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful. Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements. ### Keep evidence and independent contributions distinct Distinguish evidence, inference, and missing information. Check your work against the source material. Use independent agents when your system actually supports delegation and independent investigation or review would improve the work: - Give each agent a bounded brief and the context appropriate to its role. - Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations. - Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately. ### Produce, check, and revise the result Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them. Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result. Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized. Help me find waste in my team’s expenses by investigating the purpose, use and commitments behind actual charges. Produce an itemized review with supported amounts and practical consequences. Do not assume that expense, high cost or low visible usage establishes waste, and do not cancel, buy or change anything. ### 1. Establish the spending scope and access Start with the team, expense file, period or question I supplied. Inspect and try relevant finance, document and communication connections. Establish the currency, business units, expense categories and date range to examine. Ask one question at a time when a consequential scope choice is unclear, and wait on dependent investigation. Search expense records, invoices, card or reimbursement entries, purchase orders, subscriptions, contracts and license or service usage. Search related emails, chats including WhatsApp where accessible, calendars, meeting transcripts and project documents for why the spending was approved, what work it supported, who used it and what commitments remain. A calendar event can contextualize travel or a service; it does not alone prove use or non-use. If access fails, investigate permitted connection or export options before requesting specific files or a user-only action. Welcome rough spreadsheets or copied records, but explain early which missing invoices, credits, usage histories or contract terms prevent a reliable estimate. An interview can identify leads; it cannot establish transaction completeness or verified savings. ### 2. Reconcile the transaction record Create a working ledger preserving original row or invoice references. Record vendor, date, service period, category, amount, currency, tax, cost center, recurring status and linked contract where available. Distinguish an invoice from its payment so the same expense is not counted twice. Account for refunds, credits, reversals, reimbursements and overlapping exports. Reconcile the reviewed ledger with the available source totals and report any difference. State covered accounts and periods; do not call the review comprehensive if only a subset is accessible. Track amended agreements, renewed subscriptions and effective dates. A newer invoice may reflect changed volume, tax or a new service period rather than an unjustified price rise. Create an initial list of candidates such as duplicate charges, unused paid capacity, continuing spend for ended work, overlapping services or avoidable fees. Treat these as questions to investigate. Preserve the underlying rows so every candidate can be traced to a specific charge. ### 3. Investigate candidates in bounded categories Use agents for bounded categories, vendors or periods when useful. Give each its exact ledger subset, invoice and contract references, contextual sources, investigation questions and required deliverable. Completion means every assigned candidate has an evidence-based finding or a named missing record, and assigned rows are accounted for. Maintain a central coverage record and prevent overlapping assignments from producing duplicate savings. Each investigator should establish: - What was purchased, for whom and for which work. - Actual use, alternatives already paid for and any necessary reserve capacity. - Whether an apparent duplicate covers the same service and period or a legitimate separate obligation. - Relevant term, notice period, minimum commitment, renewal and cancellation conditions. - Evidence for the problem, plausible legitimate explanations and what would change if the expense were reduced. Use waves as needed. Without agents, use manageable sequential passes or fresh sessions with explicit source, findings and coverage handoffs. State honestly when review shares the investigating context. ### 4. Verify problems and calculate realizable savings Recheck promising findings against invoices, usage and the work record. Look for seasonal needs, backup capacity, upcoming projects, service requirements and agreements that explain apparent non-use. Ask me one focused question when a material purpose or obligation remains unresolved, and wait before classifying that candidate as verified waste. Separate confirmed billing errors or unnecessary spend from negotiable opportunity and unverified candidates. For each supported change, calculate its basis transparently: units, rate, eligible period, effective change date, fees, replacement costs and one-time effort where known. Distinguish a recoverable past charge from future avoidable expense. Show first-year and recurring effects separately when relevant; do not annualize a short observation without an explicit assumption. Avoid adding overlapping opportunities. Distinguish gross potential from net realizable savings and use a range when evidence supports uncertainty. An annual commitment may permit no immediate saving even if seats are unused. State that plainly. Zero verified waste is an acceptable result. ### 5. Review and deliver the itemized findings Give an independent reviewer the actual ledger, findings, calculations and relevant original sources without your explanation of the intended conclusion. Ask it to verify the strongest cases, arithmetic, credits, duplication, contract timing and consequences for the team’s work. Resolve material findings before delivery. Use a fresh session if possible, or identify a shared-context check. Deliver the itemized review with vendor or expense, source reference, observed issue, verified facts, remaining uncertainty, amount paid, supported recoverable or avoidable amount, timing and effect of the proposed change. Summarize verified totals separately from unresolved possibilities and list material coverage gaps. Include a short set of ready-to-use questions or draft internal requests needed to resolve the strongest remaining candidates. Do not send those requests or execute a financial change. Finish with the evidence needed for a spending decision, not a claim that projected savings have already been achieved. These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.
View full AI instructions
Find the waste hiding in your team’s expenses.
What the AI will do
- Inspect expenses, invoices, subscriptions and usage, then search messages and project records for the purpose and commitments behind the spending.
- Investigate bounded categories, verify promising findings, and use a separate reviewer where available to check the amounts and consequences.
What you can provide
- The team, time period and any known spending question.
- Access to expense records, invoices, subscriptions, usage and relevant project communications; files or usable exports also work.
Where you'll take part
- Clarify material gaps about use, commitments or service needs that the records cannot resolve.
- Review the itemized findings and choose whether any proposed change should proceed separately.
What you'll get
An itemized expense review showing verified problems, unresolved candidates, supported savings estimates and what each proposed change would affect; no purchases or cancellations are made.
Activity context
An expense is not waste just because it is large, recurring or unfamiliar. Follow the money back to the work it supported, and see which charges survive that examination.
For the user
Name the team and period, or begin with the expense file you have. The AI will investigate the records and context, verify likely problems, and ask only about material uncertainty. Missing invoices, usage or commitments can limit both the findings and savings estimate. The activity ends with a review you can act on, not automatic cancellations.
Guidance for the AI
These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result.
Start with the activity and the user's context
Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear.
Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them.
Check the capabilities needed for this work
Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit.
When useful access is missing:
- Explain what you cannot reach and how that material would help.
- Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot.
- If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response.
- Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available.
Ask focused questions and respect the answers
Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful.
Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements.
Keep evidence and independent contributions distinct
Distinguish evidence, inference, and missing information. Check your work against the source material.
Use independent agents when your system actually supports delegation and independent investigation or review would improve the work:
- Give each agent a bounded brief and the context appropriate to its role.
- Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations.
- Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately.
Produce, check, and revise the result
Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them.
Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result.
Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized.
Your AI prompt
Help me find waste in my team’s expenses by investigating the purpose, use and commitments behind actual charges. Produce an itemized review with supported amounts and practical consequences. Do not assume that expense, high cost or low visible usage establishes waste, and do not cancel, buy or change anything.
1. Establish the spending scope and access
Start with the team, expense file, period or question I supplied. Inspect and try relevant finance, document and communication connections. Establish the currency, business units, expense categories and date range to examine. Ask one question at a time when a consequential scope choice is unclear, and wait on dependent investigation.
Search expense records, invoices, card or reimbursement entries, purchase orders, subscriptions, contracts and license or service usage. Search related emails, chats including WhatsApp where accessible, calendars, meeting transcripts and project documents for why the spending was approved, what work it supported, who used it and what commitments remain. A calendar event can contextualize travel or a service; it does not alone prove use or non-use.
If access fails, investigate permitted connection or export options before requesting specific files or a user-only action. Welcome rough spreadsheets or copied records, but explain early which missing invoices, credits, usage histories or contract terms prevent a reliable estimate. An interview can identify leads; it cannot establish transaction completeness or verified savings.
2. Reconcile the transaction record
Create a working ledger preserving original row or invoice references. Record vendor, date, service period, category, amount, currency, tax, cost center, recurring status and linked contract where available. Distinguish an invoice from its payment so the same expense is not counted twice. Account for refunds, credits, reversals, reimbursements and overlapping exports.
Reconcile the reviewed ledger with the available source totals and report any difference. State covered accounts and periods; do not call the review comprehensive if only a subset is accessible. Track amended agreements, renewed subscriptions and effective dates. A newer invoice may reflect changed volume, tax or a new service period rather than an unjustified price rise.
Create an initial list of candidates such as duplicate charges, unused paid capacity, continuing spend for ended work, overlapping services or avoidable fees. Treat these as questions to investigate. Preserve the underlying rows so every candidate can be traced to a specific charge.
3. Investigate candidates in bounded categories
Use agents for bounded categories, vendors or periods when useful. Give each its exact ledger subset, invoice and contract references, contextual sources, investigation questions and required deliverable. Completion means every assigned candidate has an evidence-based finding or a named missing record, and assigned rows are accounted for. Maintain a central coverage record and prevent overlapping assignments from producing duplicate savings.
Each investigator should establish:
- What was purchased, for whom and for which work.
- Actual use, alternatives already paid for and any necessary reserve capacity.
- Whether an apparent duplicate covers the same service and period or a legitimate separate obligation.
- Relevant term, notice period, minimum commitment, renewal and cancellation conditions.
- Evidence for the problem, plausible legitimate explanations and what would change if the expense were reduced.
Use waves as needed. Without agents, use manageable sequential passes or fresh sessions with explicit source, findings and coverage handoffs. State honestly when review shares the investigating context.
4. Verify problems and calculate realizable savings
Recheck promising findings against invoices, usage and the work record. Look for seasonal needs, backup capacity, upcoming projects, service requirements and agreements that explain apparent non-use. Ask me one focused question when a material purpose or obligation remains unresolved, and wait before classifying that candidate as verified waste.
Separate confirmed billing errors or unnecessary spend from negotiable opportunity and unverified candidates. For each supported change, calculate its basis transparently: units, rate, eligible period, effective change date, fees, replacement costs and one-time effort where known. Distinguish a recoverable past charge from future avoidable expense. Show first-year and recurring effects separately when relevant; do not annualize a short observation without an explicit assumption.
Avoid adding overlapping opportunities. Distinguish gross potential from net realizable savings and use a range when evidence supports uncertainty. An annual commitment may permit no immediate saving even if seats are unused. State that plainly. Zero verified waste is an acceptable result.
5. Review and deliver the itemized findings
Give an independent reviewer the actual ledger, findings, calculations and relevant original sources without your explanation of the intended conclusion. Ask it to verify the strongest cases, arithmetic, credits, duplication, contract timing and consequences for the team’s work. Resolve material findings before delivery. Use a fresh session if possible, or identify a shared-context check.
Deliver the itemized review with vendor or expense, source reference, observed issue, verified facts, remaining uncertainty, amount paid, supported recoverable or avoidable amount, timing and effect of the proposed change. Summarize verified totals separately from unresolved possibilities and list material coverage gaps. Include a short set of ready-to-use questions or draft internal requests needed to resolve the strongest remaining candidates. Do not send those requests or execute a financial change. Finish with the evidence needed for a spending decision, not a claim that projected savings have already been achieved.
Done when
You have a checked itemized review, with verified amounts, unresolved leads, realistic savings calculations and the effects of proposed changes clearly shown. The review method and any limits are stated.
These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.
These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.